DESCRIPTION: Despite the availability of curative therapy, tuberculosis continues to be the leading cause of death from a bacterial pathogen worldwide. Effective TB control requires the early diagnosis and treatment of infectious cases to prevent further transmission of the disease. Much of the problem stems from the failure of traditional TB diagnostics which rely on sputum smear microscopy and to a lesser extent culture confirmation of disease and drug resistance phenotypes. Not only are sputum smears of low diagnostic yield, cultures for Mycobacterium tuberculosis grow notoriously slowly. Since drug resistance is traditionally assessed by culture, drug resistant TB is often detected months after the initiation of empirical drug regimens. These delays not only lead to the acquisition of further resistance in patients receiving inadequate regimens but also to the prolonged infectiousness of those with resistant organisms and the consequent spread of drug resistant strains. The recent development of rapid molecular diagnostics promises to revolutionize TB control. Such tests can be performed directly on sputum or other clinical samples and rely on the rapid detection of MTB genetic signatures as well as resistance-associated mutations. Clinical studies demonstrate that they perform better than sputum smear microscopy and reliably identify resistance to the first line drug, rifampicin. However, these tools suffer from two important limitations. First, current platforms are substantially less sensitive than culture and often miss smear negative TB. And secondly, gaps in our knowledge of the genetic determinants of phenotypic resistance limit the spectrum of drugs to which resistance can be detected. Here, we propose to address these gaps through a multi-disciplinary collaboration emphasizing discovery of new biomarkers of resistance, the identification of optimal clinical sampling strategies directed toward detection of MTB DNA and the development of a sensitive micro-array based rapid diagnostic. Our long-term goal is to develop a diagnostic strategy that will improve the diagnosis of childhood and DR TB and stem the further spread of the disease.